The Corridor Compressed 2.0 Basis Points. Every Basis Point of the 10yr's 7.0 Came Out of Inflation.

2s10s is 34.9 basis points on the 6 November UST marks, from 36.9 a week earlier, and the long end did all of the compressing: the 10yr fell 7.0 to 4.859% against the 2yr's 5.0 to 4.51%. Every basis point of that move came out of inflation compensation. The 10yr breakeven fell 11.0 to 2.23% (as of 6 November), the lowest since 5 August, while the real 10yr rose 4.0 to 2.629%. High-yield OAS went the other way: 3.48% (as of 5 November), the 2026 high, with gold up 5.0% on the week.

The corridor is the gap between what Washington pays for 2 years of money and what it pays for 10, and it is the market's one-line opinion on whether the next few years are a policy story or a term-premium story. A week ago that opinion was 36.9 basis points, set by a front end repricing the policy path. On Friday's close it is 34.9, and the engine has changed ends. The 10yr fell 7.0 basis points against the 2yr's 5.0, so the long end supplied the compression. That would normally be a duration bid, an insurance purchase. It is not. The 10yr breakeven fell 11.0 basis points to 2.23% while the real yield rose 4.0 to 2.629%: the bond market bought less inflation, not easier money, and it charged 4 basis points more in real terms for the privilege. Credit did not join. High-yield OAS is 3.48%, the widest print of 2026, and gold is up 5.0% with real yields higher. Three markets, three stories, one week.

What moved

The 2yr closed at 4.51% and the 5yr at 4.64%. The 7yr 4.74%, the 10yr 4.859%, the 20yr 5.29%, the 30yr 5.26%. The 3-month bill 4.19%. All levels are the 6 November close; the snapshot was taken at 07:00 UTC today.

Every tenor from the 2yr to the 30yr fell; only the bill rose. The 2yr took 5.0 basis points, the 5yr 7.0, the 7yr 6.0, the 10yr 7.0, the 20yr 5.0, the 30yr 5.0. The 3-month bill added 3.0 to 4.19%, the only point on the curve to move higher. A bill that rises while the 2yr falls is the front end of the front end saying the near-term cash rate is not going anywhere.

The back end is still the expensive part of the curve. The 30yr sits at the 76th percentile of its 3yr range, the 20yr at the 75th, the 10yr at the 73rd, the 5yr and 7yr at the 73rd, the 2yr at the 68th and the 3-month bill at the 30th. Nothing here is cheap. The curve has moved down its own range together.

The dollar was 102.063, up 0.237 on the week and at the 43rd percentile. Dollar-yen 158.126, up 0.373. A firm dollar and a rallying long end is the pairing that decides whether this week's move is a growth story or a debasement story.

The long end did the compressing

2s10s is 34.9 basis points on the UST cash marks for 6 November, from 36.9 on 30 October. That is 2.0 basis points of compression, and it is the third direction in three weeks: 29.8 on 23 October, 36.9 on 30 October, 34.9 now.

The arithmetic runs off the two marks the spread is computed from. The 2yr went from 4.56% to 4.51%, down 5.0 basis points. The 10yr went from 4.929% to 4.859%, down 7.0. 7.0 minus 5.0 is 2.0. The long end supplied the compression, which is the reverse of last Monday, when the front end supplied 7.1 basis points of widening.

The par curve agrees, once the dates are matched. Its newest row is 11/05/2026: 2yr 4.46% and 10yr 4.84%, a 38.0 basis point spread, from 32.0 on 11/04. The par file runs a session behind the cash marks, and on the 5 November cash marks the spread was 37.7. The two measures are 0.3 basis points apart. No disagreement to explain.

The path matters more than the endpoints. 36.9 on 30 October, 30.4 on 2 November, 34.2 on 3 November, 31.4 on 4 November, 37.7 on 5 November, 34.9 at the close. A 7.3 basis point range in four sessions is not a trend; it is a market that cannot decide whether the next few years are a policy story or a term-premium story, and has been charging and refunding the same insurance all week.

The 5 November mark is the one to keep. The corridor reached 37.7 with the 2yr at 4.46% and the 10yr at 4.837%. The next session both yields rose: the 2yr +5.0 basis points, the 10yr +2.2. The spread compressed 2.8 on a selloff. A curve that flattens while both yields rise is not a duration bid; it is the near end being marked up harder than the far end.

Every basis point came out of the inflation leg

The 10yr nominal fell 7.0 basis points. Inflation compensation fell 11.0. The real yield rose 4.0.

The breakeven is 2.23% (as of 6 November), from 2.34% (as of 30 October). It is the lowest print since 5 August, when it closed at 2.22%; nothing between then and now has been this low. The 10yr real is 2.629%, from 2.589%. The Treasury's own real curve puts the 5 November print at 2.60%, which is where the same arithmetic lands a session earlier.

That is a specific kind of rally. The bond market did not buy a lower policy path through the long end; it bought less inflation compensation and demanded 4.0 basis points more real yield for the privilege. The whole of the 10yr's 7.0 basis points, and 4.0 besides, is the breakeven.

Gold disagrees. Gold is $4,121.90, up $197.90 on the week, 5.0%, with the real 10yr 4.0 basis points higher. The usual relationship runs the other way: a higher real yield raises the carry cost of a metal that pays nothing. Silver did the same, up 4.8% to $52.92. Two inflation hedges rallying into a collapse in inflation compensation is the loudest disagreement on this tape.

Crude explains part of the breakeven and none of the gold. WTI is $89.46, down 11.0% on the week; Brent $97.10, down 11.6%. Lower crude pulls expected headline inflation down, which is half of the breakeven story. Copper fell 3.2% to $6.3215 and the copper-gold ratio is 0.00154, from 0.00166, so the industrial metal lost ground to the monetary one. Crude lower and gold higher is a demand signal and a currency signal running at the same time.

Credit went the other way

High-yield OAS is 3.48% (as of 5 November), from 3.32% (as of 4 November) and 3.26% (as of 29 October). That is 22 basis points wider in the week to 5 November, the highest print in this file since 9 May 2025, when it closed at 3.53%, and the 2026 high.

Investment-grade OAS is 0.93% (as of 5 November), from 0.89%. It sits 1 basis point under the 0.94% high, last touched on 2 November, 30 October and 16 March. High yield has moved; investment grade has not.

The cash wrappers were quieter than the spreads. HYG fell $0.20 to $77.23 and JNK $0.28 to $92.70. LQD added $0.26 to $104.53 and TLT $0.08 to $81.92.

Equities repriced. The S&P 500 is 7,376.86, down 203.21 on the week, 2.7%; the Nasdaq 100 26,989.99, down 1,394.86, 4.9%; QQQ $656.10, down $34.39.

The two volatility complexes separated. The VIX is 25.64, up 6.22 points on the week and at the 34th percentile. The MOVE is 92.77, down 7.43, at the 44th. Rate volatility fell 7.43 points while default insurance went to a 2026 high. Whatever is being priced this week is not being priced in rates.

A breakeven 11 basis points lower with high yield at its widest of the year is a market pricing less inflation and more default at the same time. Those are not the same trade, and they do not usually sit at the same address. Credit spreads widen before defaults arrive, not after, which makes the credit leg the earlier warning of the two.

The drain paused for a week

Reserve balances were $2,930.1bn in the week ended 4 November, from $2,893.8bn on 28 October: $36.3bn added back, after $68.8bn came out the week before.

The Treasury's general account was $922.8bn on 5 November, from $951.6bn on 30 October, $28.8bn lower, and $129.7bn below the 27 October peak of $1,052.5bn. A falling chequebook is a reserve injection, which is where the reserves came from.

Overnight RRP take-up was $0.15bn on 6 November, from $4.375bn on 30 October and $5.087bn on 2 November. The month-end fill drained away again, which is what that facility does now: one day of use at the turn, then back to rounding error.

Secured overnight money fixed 3.92% on 5 November against an administered 3.90%: a 2 basis point premium, from 7 on 3 November and 4 on 4 November. Effective fed funds is 3.88%. The premium is decaying as the calendar normalises, which is the benign version of a funding market under strain.

The composite is 0.421, TIGHTENING. Scarcity is dominant at 0.945, Credit & Stress 0.907, Rates & Policy 0.652, Growth & Inflation -0.037, Liquidity -0.363. Liquidity is the only negative sub-index, and it has turned from +0.005 on 2 November.

Claims are 189,000 in the week to 31 October, from 195,000. Continuing claims 1,718,000 in the week to 24 October, from 1,695,000. Core CPI 2.57% and headline 3.72%, both September, neither refreshed since.

Bitcoin: the discount narrowed while the coin fell

BTC is $74,228.36 at 07:00 UTC, down $4,034.21 on the week, 5.2%, and 10.3% over thirty sessions. ETH $2,110.64, down 7.8%.

MSTR is $128.93, down $1.58. On 846,000 coins and 384.23m shares, that is $163.42 of coins per share, so the equity trades at 0.789 times its holdings: a 21.1% discount. On 30 October the same arithmetic on $130.51 over $172.32 was 0.757. The discount narrowed.

The preferreds firmed. STRC $99.84, up $0.71, at the 100th percentile of its range. STRF $103.20, up $1.78. STRK $67.85, unchanged on the week at the 29th percentile.

A coin down 5.2% and a wrapper down 1.2% is the multiple doing the work in the other direction from last week, when the coin fell 3.7% and the wrapper fell 10.6%. The real 10yr is 4.0 basis points higher on the week, so carry got more expensive and the discount narrowed anyway. Cheap carry was never the only thing holding the structure up, and this week it was not the thing at all.

What would change the read

The first is the pair holding. Breakevens at or below 2.23% with high-yield OAS through 3.53%, the May 2025 print, would move this from a disinflation repricing to a growth break. That is the combination both markets are pointing at, and it is the one that makes the equity move the leading one rather than the echo.

The second is the inflation leg turning back. A breakeven above 2.34% with the 10yr still under 4.859% puts the whole move in the real leg, which is the policy-easing read, and it is the read the front end has been carrying since the 28 October repricing.

The third is credit. High-yield OAS back under 3.32%, the 4 November print, with the 10yr still under 4.86%, would say the widening was a week of supply rather than a change in default expectations, and the curve has it right.

Between those three lines the read is narrow: the corridor compressed 2.0 basis points because the long end fell 7.0 to the front end's 5.0, and all of that 7.0 was inflation compensation, 11.0 of it, against 4.0 of real yield added. Credit went to a 2026 high in the same week. The curve is pricing less inflation. The credit market is pricing more default. One of them is early.